By Matthew Rayback, StorageCraft
I think most business owners would agree in theory that disaster recovery is important. In practice, however, it’s a little too easy to cut corners. There are so many options, it’s easy to settle on something simple and cheap. But if you haven’t taken the time to understand the kind of disaster recovery solution your business needs, you could be missing something big and paying for a solution that will let you down in the end.
I asked six managed service providers what, in their experience, was the most overlooked aspect of any disaster recovery solution, because they’d know. Between them, they work with hundreds of businesses, all of whom are in various stages of denial about just how much disaster recovery they need.
To be honest, I was a little surprised by their answers. I spend a lot of time looking at statistics that try to predict the future of the IT industry and that makes it easy to forget that in the middle of the cloud, complex virtualization platforms, RMMs, and XaaS, a lot of people are still just trying to figure out the basics.
So what are you overlooking in your disaster recovery solution? Let’s see what the experts have to say.
1. Recovery (Wait, what?)
When talking about disaster recovery, it seems weird that you might overlook the whole recovery part of it, but according to John Motezedi, the CEO of SNC Squared, it’s a big problem. In response to my email, he answered simply “restoration, testing, and recovery of data.” (We’ll talk about testing in a bit.)
SNC Squared is no stranger to recovery. After a titanic tornado tore through its home town of Joplin, Missouri in 2011, SNC Squared headquarters lay in ruins along with those of a huge number of its clients. But that didn’t stop John or his team. They recovered 100 percent of their customers’ data within seventy-two hours. So when John talks about the importance of the recovery, not just the backup, he knows what he’s talking about.
It’s easy to think only about the backup. You put your solution in place, let it do its thing, and forget about it. People often compare disaster recovery with an insurance policy, where you pay your premiums and you forget about it.
But with disaster recovery, paying your premium by making backups is only part of the picture. If you’re not prepared to use those backups in an actual recovery, you’re sort of missing the point.
This is especially true for businesses with legacy environments. For example, a business that’s still using tape, is a business that’s probably ignoring the recovery. In spite of its other (debatable) benefits, tape simply can’t cut it in a real recovery.
“After Sandy hit, if you’d brought us a tape, our first reaction would be to throw it back at you. We didn’t have time to build all new servers and restore them via tape. We didn’t have a hundred extra man hours in the middle of a crisis.”
When it comes down to it, all the advice I received from the managed service providers I talked to had this at its core. If you’re not thinking past the backup, you’re going to regret it.
2. Good backups
But that doesn’t mean you can forget about the backup either. In fact, the road to recovery begins with the right kind of backup.
According to Teresa Bell, president of Citrus Networking Solutions Group, a proper backup is the most overlooked feature of disaster recovery. Specifically, “imaging your entire server, not just your files.”
Today, eliminating downtime (as opposed to simply recovering data) is a huge piece of the disaster recovery puzzle, and file-and-folder solutions simply cannot affect quick recoveries. The ability to reinstall your operating system, settings, applications, files, and so on from a single backup image makes recovery significantly easier.
At the same time, you need to be able to access those backup images no matter what, and so Teresa also recommends offsite backup. You always want a local backup, if for no other reason than efficiency, but if you’re overlooking an offsite backup, you’re going to be in trouble if something bigger than a server failure gets you.
“Too many apps out there talk about the great features they offer, all the bells and whistles, but they fail on the most simple of tasks. A good example of this is some vendors who use VSS with Microsoft. VSS in previous server versions, like Server 2003 (still around in great numbers), caused many issues with backups and the time it took to resolve this was too long. Our current vendor has a mechanism that if VSS fails, will do a regular backup without VSS. So you may not have the image recovery capability, but you have the data. For me, a product that actually works successfully more times than it fails is the best feature available.”
According to these IT professionals, the goal is recovery, but the path is backup and if you’re not looking at how you’re backing up, what you’re backing up, and where you’re backing up to, you’re asking for trouble.
In the end, however, most of the experts I talked to felt that the thing most businesses were missing landed somewhere in between.
As John Motazedi already told us, testing is crucial.
You may have the best backup software in the world, but if you’re not testing those backups on a regular basis, you have no idea what they’re going to do in a recovery situation.
According to Joe Hillis, the operations director for the awesome Information Technology Disaster Resource Center, the verification of recovery is a crucial task in a disaster recovery solution.
“Although there may be no error messages generated during the backup process,” he says, “the only way to know for sure is to actually perform a recovery. In complex environments, something as simple as a DNS entry or drive mapping can cause a recovery to fail, but you may never know it until you test.”
Curtis Flax from Nex-Tech agrees. Testing and verification are essential.
“I’ve seen it more times than I’d like. A customer calls us needing a restore, but they hadn’t been notified that their backups were faulty and their data is lost.”
Of course, testing comes in many shapes and a sizes, from simple verifications to full-on test recoveries, but one way or another, they are the key to a successful recovery and unfortunately, many businesses aren’t doing them.
“Testing disaster recovery can be time consuming,” Joe says. “It may even impact production environments if not performed correctly. However, one can only assume they can recover a server until they’ve actually done it.”
“It’s a simple setup,” Curtis adds, “and most overlook it.”
Taking a Little Extra Time
Disaster recovery is a big deal. We’re talking about your business here. You don’t want to cut corners, you want to do it right. And while it may be true that “it’s typically human behavior that keeps a backup and disaster recovery solution from working properly,” as Joe says, that human behavior shouldn’t be your negligence.
So, when planning for your disaster recovery, take a little extra time at the beginning to make sure you’ve got everything you need in your solution, and you’ll save your business in the end.